The position of the EU Council on the flagship legislation to regulate Artificial Intelligence was shared on Friday with some final last-minute adjustments made by the Czech Presidency.
The text has been sent to the other member states to obtain the go-ahead from EU ambassadors.
“The Permanent Representatives Committee is invited to examine and approve the text set out in the Annex to this note, with a view to a general approach; recommend that the Council, at the meeting of the TTE (Telecommunications) Council on 6 December 2022, reach a general approach,” the document reads.
Read more: EU Council Nears Common Position On AI Act
In the EU Council, a heated topic was which AI applications in the insurance field should be deemed at risk of causing considerable harm. In a final tweak to the text, only algorithms used for the risk assessment and pricing of health and life insurance are considered high risk. In contrast, the rest remains covered by sectoral legislation.
Algorithms used to evaluate individuals’ credit scores, creditworthiness or insurance premiums were put in the high-risk basket, considering this a vital way for people to access essential services. However, micro or small companies putting into service these systems for their own use have been exempted.
The latest text clarifies what ‘their own use’ means, namely that the micro and small enterprises “put into service an AI system for the purpose of selling their own insurance products” in case of the insurance high-risk use case. The same rules would respectively apply to credit institutions.